Computing systems and methods for identifying and providing information about recurring transactions

Abstract

An example computing platform is configured to: (a) obtain data for a given transaction involving a given customer account and a given merchant, (b) apply pre-processing logic to the obtained data for the given transaction and thereby derive feature data for the given transaction, (c) input the feature data for the given transaction into a trained machine-learning model that functions to (i) evaluate the feature data for the given transaction and (ii) based on the evaluation, output a score for the given transaction that indicates a likelihood that the given transaction is a recurring charge, and (d) based on the score for the given transaction, determine whether to classify the given transaction as a recurring charge.

1 . A computing platform comprising:
 a network interface;   at least one processor;   at least one non-transitory computer-readable medium; and   program instructions stored on the at least one non-transitory computer-readable medium that, when executed by the at least one processor, cause the computing platform to:
 for each respective financial transaction of a plurality of financial transactions involving a given customer account and a respective merchant:
 derive, for the respective financial transaction, a respective set of feature data comprising values for account-merchant-level features that provide information about past transaction history between the given customer account and the respective merchant, wherein the account-merchant-level features include at least: (i) an indication of an extent of past financial transactions involving both the given customer account and the respective merchant that follow a first type of frequency-based transaction pattern and (ii) an indication of an extent of past financial transactions involving both the given customer account and the respective merchant that follow a second type of frequency-based pattern, wherein the second type of frequency-based pattern differs from the first type of frequency-based transaction pattern; and 
 utilize a trained-machine learning model to evaluate the respective set of feature data and thereby produce a respective prediction of whether the respective financial transaction is a recurring charge; and 
 
 utilize the respective predictions of whether the respective financial transactions are recurring charges as a basis for causing an account holder of the given customer account to be presented with one or both of (i) notifications of recurring charges involving the given customer account or (ii) a recurring-charges dashboard comprising a listing of recurring charges involving the given customer account. 
   
     
     
         2 . The computing platform of  claim 1 , wherein, for each respective financial transaction of the plurality of financial transactions, the respective set of feature data further comprises a value of at least one transaction-level feature that provides information about the respective financial transaction. 
     
     
         3 . The computing platform of  claim 2 , wherein the at least one transaction-level feature comprises at least one of (i) an indication of whether a transaction time of the respective financial transaction was within a defined window of time or (ii) an indication of whether a transaction amount of the respective financial transaction was within a defined range of amounts. 
     
     
         4 . The computing platform of  claim 1 , wherein, for each respective financial transaction of the plurality of financial transactions, the respective set of feature data further comprises a value of at least one merchant-level feature that provides information about the respective merchant involved in the respective financial transaction. 
     
     
         5 . The computing platform of  claim 4 , wherein the at least one merchant-level feature comprises at least one of (i) an indication of a size of the respective merchant, (ii) an indication of a concentration of the respective merchant's past financial transactions with respect to transaction dates, transaction times, or transaction amounts, or (iii) an indication of an extent of the respective merchant's past financial transactions that share at least one common characteristic with the respective financial transaction. 
     
     
         6 . The computing platform of  claim 1 , wherein the first type of frequency-based transaction pattern and the second type of frequency-based transaction pattern comprise first and second ones of a weekly transaction pattern, a bi-weekly transaction pattern, a monthly transaction pattern, a bi-monthly transaction pattern, a quarterly transaction pattern, a bi-annual transaction pattern, or an annual transaction pattern. 
     
     
         7 . The computing platform of  claim 1 , wherein the respective predictions of whether the respective financial transactions are recurring charges are utilized as a basis for causing the account holder of the given customer account to be presented with the notifications of recurring charges involving the given customer account, and wherein each of the notifications comprises one of a mobile push notification, an email notification, or a text-message notification that is presented via a networked end-user device associated with the account holder. 
     
     
         8 . The computing platform of  claim 7 , wherein at least one of the notifications includes a selectable element for accessing a recurring-charges dashboard. 
     
     
         9 . The computing platform of  claim 1 , wherein the respective predictions of whether the respective financial transactions are recurring charges are utilized as a basis for causing the account holder of the given customer account to be presented with the recurring-charges dashboard comprising the listing of recurring charges involving the given customer account, and wherein each respective recurring charge is associated with a selectable element that, when selected, causes the recurring-charges dashboard to present any other financial transaction involving the given customer account and a same merchant as the listed recurring charge that is determined to be related to the listed recurring charge. 
     
     
         10 . The computing platform of  claim 1 , further comprising program instructions stored on the at least one non-transitory computer-readable medium that are executable by the at least one processor such that the computing platform is configured to:
 for each respective financial transaction that is predicted to be a recurring charge, update stored transaction-level data for the respective financial transaction to include additional data that enhances the stored transaction-level data.   
     
     
         11 . The computing platform of  claim 1 , wherein the plurality of financial transactions comprise card-not-present transactions involving the given customer account. 
     
     
         12 . At least one non-transitory computer-readable medium, wherein the at least one non-transitory computer-readable medium is provisioned with program instructions that, when executed by at least one processor cause a computing platform to:
 for each respective financial transaction of a plurality of financial transactions involving a given customer account and a respective merchant:
 derive, for the respective financial transaction, a respective set of feature data comprising values for account-merchant-level features that provide information about past transaction history between the given customer account and the respective merchant, wherein the account-merchant-level features include at least: (i) an indication of an extent of past financial transactions involving both the given customer account and the respective merchant that follow a first type of frequency-based transaction pattern and (ii) an indication of an extent of past financial transactions involving both the given customer account and the respective merchant that follow a second type of frequency-based pattern, wherein the second type of frequency-based pattern differs from the first type of frequency-based transaction pattern; and 
 utilize a trained-machine learning model to evaluate the respective set of feature data and thereby produce a respective prediction of whether the respective financial transaction is a recurring charge; and 
   utilize the respective predictions of whether the respective financial transactions are recurring charges as a basis for causing an account holder of the given customer account to be presented with one or both of (i) notifications of recurring charges involving the given customer account or (ii) a recurring-charges dashboard comprising a listing of recurring charges involving the given customer account.   
     
     
         13 . A method carried out by a computing platform, the method comprising:
 for each respective financial transaction of a plurality of financial transactions involving a given customer account and a respective merchant:
 deriving, for the respective financial transaction, a respective set of feature data comprising values for account-merchant-level features that provide information about past transaction history between the given customer account and the respective merchant, wherein the account-merchant-level features include at least: (i) an indication of an extent of past financial transactions involving both the given customer account and the respective merchant that follow a first type of frequency-based transaction pattern and (ii) an indication of an extent of past financial transactions involving both the given customer account and the respective merchant that follow a second type of frequency-based pattern, wherein the second type of frequency-based pattern differs from the first type of frequency-based transaction pattern; and 
 utilizing a trained-machine learning model to evaluate the respective set of feature data and thereby produce a respective prediction of whether the respective financial transaction is a recurring charge; and 
   utilize the respective predictions of whether the respective financial transactions are recurring charges as a basis for causing an account holder of the given customer account to be presented with one or both of (i) notifications of recurring charges involving the given customer account or (ii) a recurring-charges dashboard comprising a listing of recurring charges involving the given customer account.   
     
     
         14 . The method of  claim 13 , wherein, for each respective financial transaction of the plurality of financial transactions, the respective set of feature data further comprises a value of at least one transaction-level feature that provides information about the respective financial transaction. 
     
     
         15 . The method of  claim 13 , wherein, for each respective financial transaction of the plurality of financial transactions, the respective set of feature data further comprises a value of at least one merchant-level feature that provides information about the respective merchant involved in the respective financial transaction. 
     
     
         16 . The method of  claim 13 , wherein the first type of frequency-based transaction pattern and the second type of frequency-based transaction pattern comprise first and second ones of a weekly transaction pattern, a bi-weekly transaction pattern, a monthly transaction pattern, a bi-monthly transaction pattern, a quarterly transaction pattern, a bi-annual transaction pattern, or an annual transaction pattern. 
     
     
         17 . The method of  claim 13 , wherein the respective predictions of whether the respective financial transactions are recurring charges are utilized as a basis for causing the account holder of the given customer account to be presented with the notifications of recurring charges involving the given customer account, and wherein each of the notifications comprises one of a mobile push notification, an email notification, or a text-message notification that is presented via a networked end-user device associated with the account holder. 
     
     
         18 . The method of  claim 13 , wherein the respective predictions of whether the respective financial transactions are recurring charges are utilized as a basis for causing the account holder of the given customer account to be presented with the recurring-charges dashboard comprising the listing of recurring charges involving the given customer account, and wherein each respective recurring charge is associated with a selectable element that, when selected, causes the recurring-charges dashboard to present any other financial transaction involving the given customer account and a same merchant as the listed recurring charge that is determined to be related to the listed recurring charge. 
     
     
         19 . The method of  claim 13 , further comprising:
 for each respective financial transaction that is predicted to be a recurring charge, update stored transaction-level data for the respective financial transaction to include additional data that enhances the stored transaction-level data.   
     
     
         20 . The method of  claim 13 , wherein the plurality of financial transactions comprise card-not-present transactions involving the given customer account.